Comments on: The Procedural Generation.

By: Irun

Irun — Tue, 07 Jul 2015 14:54:55 +0000

Uhm… Well, sorry for that necro. ‘:]

By: Irun

Irun — Tue, 07 Jul 2015 14:31:41 +0000

I’m quite sure every problem you noticed about PGC here, can also be found in purely human crafted content. The true issue is that, still today, in a game project employing 200, there will be 75 dedicated to graphics and 1 or 2 people for PGC. So, I don’t think that the games we can try out there are in any way representative of what it should(IMO) be. I’m always depressed when I see that the most recent and best procedural content-based games are so often the results of only one passionate developer. Imagine what they could actually achieve being 75 on this problem… it could be polished, optimized, realist, tested, etc…

Yes, I really hope large studios will change the way they manage to produce (good?) titles. I personally couldn’t care less about the opus 25th with 4k more polygons and the very last particle effect.
Unfortunately, this directly reflect to the society. Most of the new gamers will neglect gameplay and choose graphics, as they would do on a regular day, choosing a fast-food over a fine restaurant (One is more appealing, the other more complicated). They will abandon replayability and buy the updated version next year, as they do with their iPhones in the consumer society.

Finally, after all those general trends, I still keep hope for very good games to appear in the future. Because big studios have one main goal: the profit (it’s normal, I won’t blame them for this), and they will be inspired by bigger successes of smaller teams when noticing their loss of income.

By: Blah

Blah — Sat, 21 Dec 2013 01:11:27 +0000

Not quite practical for small studios or individuals working on games, but I get the sentiment.

Lets not forget that TES II was procedural.

By: Michael Cook

Michael Cook — Sun, 21 Oct 2012 14:20:33 +0000

I agree with a lot of your examples, but I think the problems with PCG’s use just highlight how important it is that more developers actually try to innovate with it, rather than using it as a patch to cover a content problem.

Also, world generation is tricky, and we’re not really that good at it – but we don’t try to PCG other things at all. Another reason to fund better game AI research, rather than splurging over SIGGRAPH so we can get shinier eyeballs in CryEngine 5.

By: Hentzau

Hentzau — Sat, 29 Sep 2012 11:57:31 +0000

I’m sure they could do better, but they would literally have to chuck out the entire game and start over. There’s too much weird stuff like the fluid system hardcoded into it that precludes realistic water features. It’d be interesting to see a Minecraft 2 that tackled these problems from the ground up, though.

Also I really need to go back to the server and check out the changes.

By: Gap

Gap — Sat, 29 Sep 2012 00:53:52 +0000

It would be fun to have a Minecraft terrain generator that had more realistic geography – proper mountain ranges, rivers that flow from the hills to the sea, better view distances for appreciating vistas, etc. I suspect the lack of this is more down to the incremental nature of Minecraft development, and as it is the world works as a very simple facsimile of reality, but it’s often disconcerting to see rivers running into mountainsides or jungles butting up against tundra.

Also, on the subject of railways, the next Minecraft update apparently allows you to send minecarts through nether portals.